Radiomics in stratification of pancreatic cystic lesions: Machine learning in action

Metabolism score and machine learning models for the prediction of esophageal squamous cell carcinoma progression

The incomplete prediction of prognosis in esophageal squamous cell carcinoma (ESCC) patients is attributed to various therapeutic interventions and complex prognostic factors. Consequently, there is a pressing demand for enhanced predictive biomarkers that can facilitate clinical management and treatment decisions. This study recruited 491 ESCC patients who underwent surgical treatment at Huashan Hospital, Fudan University. We incorporated 14 blood metabolic indicators and identified independent prognostic indicators for overall survival through univariate and multivariate analyses. Subsequently, a metabolism score formula was established based on the biochemical markers. We constructed a nomogram and machine learning models utilizing the metabolism score and clinically significant prognostic features, followed by an evaluation of their predictive accuracy and performance. We identified alkaline phosphatase, free fatty acids, homocysteine, lactate dehydrogenase, and triglycerides as independent prognostic indicators for ESCC. Subsequently, based on these five indicators, we established a metabolism score that serves as an independent prognostic factor in ESCC patients. By utilizing this metabolism score in conjunction with clinical features, a nomogram can precisely predict the prognosis of ESCC patients, achieving an area under the curve (AUC) of 0.89. The random forest (RF) model showed superior predictive ability (AUC = 0.90, accuracy = 86%, Matthews correlation coefficient = 0.55). Finally, we used an RF model with optimal performance to establish an online predictive tool. The metabolism score developed in this study serves as an independent prognostic indicator for ESCC patients.

Prediction of transformation in the histopathological growth pattern of colorectal liver metastases after chemotherapy using CT-based radiomics

Chemotherapy alters the prognostic biomarker histopathological growth pattern (HGP) phenotype in colorectal liver metastases (CRLMs) patients. We aimed to develop a CT-based radiomics model to predict the transformation of the HGP phenotype after chemotherapy. This study included 181 patients with 298 CRLMs who underwent preoperative contrast-enhanced CT followed by partial hepatectomy between January 2007 and July 2022 at two institutions. HGPs were categorized as pure desmoplastic HGP (pdHGP) or non-pdHGP. The samples were allocated to training, internal validation, and external validation cohorts comprising 153, 65, and 29 CRLMs, respectively. Radiomics analysis was performed on pre-enhanced, arterial phase, portal venous phase (PVP), and fused images. The model was used to predict prechemotherapy HGPs in 112 CRLMs, and HGP transformation was analysed by comparing these findings with postchemotherapy HGPs determined pathologically. The prevalence of pdHGP was 19.8% (23/116) and 45.8% (70/153) in chemonaïve and postchemotherapy patients, respectively (P < 0.001). The PVP radiomics signature showed good performance in distinguishing pdHGP from non-pdHGPs (AUCs of 0.906, 0.877, and 0.805 in the training, internal validation, and external validation cohorts, respectively). The prevalence of prechemotherapy pdHGP predicted by the radiomics model was 33.0% (37/112), and the prevalence of postchemotherapy pdHGP according to the pathological analysis was 47.3% (53/112; P = 0.029). The transformation of HGP was bidirectional, with 15.2% (17/112) of CRLMs transforming from prechemotherapy pdHGP to postchemotherapy non-pdHGP and 30.4% (34/112) transforming from prechemotherapy non-pdHGP to postchemotherapy pdHGP (P = 0.005). CT-based radiomics method can be used to effectively predict the HGP transformation in chemotherapy-treated CRLM patients, thereby providing a basis for treatment decisions.

Delta Radiomic Features Predict Resection Margin Status and Overall Survival in Neoadjuvant-Treated Pancreatic Cancer Patients

BACKGROUND

Neoadjuvant therapy (NAT) emerged as the standard of care for patients with pancreatic ductal adenocarcinoma (PDAC) who undergo surgery; however, surgery is morbid, and tools to predict resection margin status (RMS) and prognosis in the preoperative setting are needed. Radiomic models, specifically delta radiomic features (DRFs), may provide insight into treatment dynamics to improve preoperative predictions.

METHODS

We retrospectively collected clinical, pathological, and surgical data (patients with resectable, borderline, locally advanced, and metastatic disease), and pre/post-NAT contrast-enhanced computed tomography (CT) scans from PDAC patients at the University of Texas Southwestern Medical Center (UTSW; discovery) and Humanitas Hospital (validation cohort). Gross tumor volume was contoured from CT scans, and 257 radiomics features were extracted. DRFs were calculated by direct subtraction of pre/post-NAT radiomic features. Cox proportional models and binary prediction models, including/excluding clinical variables, were constructed to predict overall survival (OS), disease-free survival (DFS), and RMS.

RESULTS

The discovery and validation cohorts comprised 58 and 31 patients, respectively. Both cohorts had similar clinical characteristics, apart from differences in NAT (FOLFIRINOX vs. gemcitabine/nab-paclitaxel; p < 0.05) and type of surgery resections (pancreatoduodenectomy, distal or total pancreatectomy; p < 0.05). The model that combined clinical variables (pre-NAT carbohydrate antigen (CA) 19-9, the change in CA19-9 after NAT (∆CA19-9), and resectability status) and DRFs outperformed the clinical feature-based models and other radiomics feature-based models in predicting OS (UTSW: 0.73; Humanitas: 0.66), DFS (UTSW: 0.75; Humanitas: 0.64), and RMS (UTSW 0.73; Humanitas: 0.69).

CONCLUSIONS

Our externally validated, predictive/prognostic delta-radiomics models, which incorporate clinical variables, show promise in predicting the risk of predicting RMS in NAT-treated PDAC patients and their OS or DFS.

Radiomics for the prediction of a postoperative pancreatic fistula following a pancreatoduodenectomy: A systematic review and radiomic score quality assessment

BACKGROUND

Postoperative pancreatic fistula (POPF) is a severe complication following a pancreatoduodenectomy. An accurate prediction of POPF could assist the surgeon in offering tailor-made treatment decisions. The use of radiomic features has been introduced to predict POPF. A systematic review was conducted to evaluate the performance of models predicting POPF using radiomic features and to systematically evaluate the methodological quality.

METHODS

Studies with patients undergoing a pancreatoduodenectomy and radiomics analysis on computed tomography or magnetic resonance imaging were included. Methodological quality was assessed using the Radiomics Quality Score (RQS) and Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) statement.

RESULTS

Seven studies were included in this systematic review, comprising 1300 patients, of whom 364 patients (28 %) developed POPF. The area under the curve (AUC) of the included studies ranged from 0.76 to 0.95. Only one study externally validated the model, showing an AUC of 0.89 on this dataset. Overall adherence to the RQS (31 %) and TRIPOD guidelines (54 %) was poor.

CONCLUSION

This systematic review showed that high predictive power was reported of studies using radiomic features to predict POPF. However, the quality of most studies was poor. Future studies need to standardize the methodology.

REGISTRATION

not registered.

An integrated machine learning-based model for joint diagnosis of ovarian cancer with multiple test indicators

OBJECTIVE

To construct a machine learning diagnostic model integrating feature dimensionality reduction techniques and artificial neural network classifiers to develop the value of clinical routine blood indexes for the auxiliary diagnosis of ovarian cancer.

METHODS

Patients with ovarian cancer clearly diagnosed in our hospital were collected as a case group (n = 185), and three groups of patients with other malignant otolaryngology tumors (n = 138), patients with benign otolaryngology diseases (n = 339) and those with normal physical examination (n = 92) were used as an overall control group. In this paper, a fully automated segmentation network for magnetic resonance images of ovarian cancer is proposed to improve the reproducibility of tumor segmentation results while effectively reducing the burden on radiologists. A pre-trained Res Net50 is used to the three edge output modules are fused to obtain the final segmentation results. The segmentation results of the proposed network architecture are compared with the segmentation results of the U-net based network architecture and the effect of different loss functions and region of interest sizes on the segmentation performance of the proposed network is analyzed.

RESULTS

The average Dice similarity coefficient, average sensitivity, average specificity (specificity) and average hausdorff distance of the proposed network segmentation results reached 83.62%, 89.11%, 96.37% and 8.50, respectively, which were better than the U-net based segmentation method. For ROIs containing tumor tissue, the smaller the size, the better the segmentation effect. Several loss functions do not differ much. The area under the ROC curve of the machine learning diagnostic model reached 0.948, with a sensitivity of 91.9% and a specificity of 86.9%, and its diagnostic efficacy was significantly better than that of the traditional way of detecting CA125 alone. The model was able to accurately diagnose ovarian cancer of different disease stages and showed certain discriminative ability for ovarian cancer in all three control subgroups.

CONCLUSION

Using machine learning to integrate multiple conventional test indicators can effectively improve the diagnostic efficacy of ovarian cancer, which provides a new idea for the intelligent auxiliary diagnosis of ovarian cancer.

Radiomics model versus 2017 revised international consensus guidelines for predicting malignant intraductal papillary mucinous neoplasms

OBJECTIVES

To evaluate a CT-based radiomics model for identifying malignant pancreatic intraductal papillary mucinous neoplasms (IPMNs) and compare its performance with the 2017 international consensus guidelines (ICGs).

MATERIALS AND METHODS

We retrospectively included 194 consecutive patients who underwent surgical resection of pancreatic IPMNs between January 2008 and December 2020. Surgical histopathology was the reference standard for diagnosing malignancy. Using radiomics features from preoperative contrast-enhanced CT, a radiomics model was built with the least absolute shrinkage and selection operator by a five-fold cross-validation. CT and MR images were independently reviewed based on the 2017 ICGs by two abdominal radiologists, and the performances of the 2017 ICGs and radiomics model were compared. The areas under the curve (AUCs) were compared using the DeLong method.

RESULTS

A total of 194 patients with pancreatic IPMNs (benign, 83 [43%]; malignant, 111 [57%]) were chronologically divided into training (n = 141; age, 65 ± 8.6 years; 88 males) and validation sets (n = 53; age, 66 ± 9.7 years; 31 males). There was no statistically significant difference in the diagnostic performance of the 2017 ICGs between CT and MRI (AUC, 0.71 vs. 0.71; p = 0.93) with excellent intermodality agreement (k = 0.86). In the validation set, the CT radiomics model had higher AUC (0.85 vs. 0.71; p = 0.038), specificity (84.6% vs. 61.5%; p = 0.041), and positive predictive value (84.0% vs. 66.7%; p = 0.044) than the 2017 ICGs.

CONCLUSION

The CT radiomics model exhibited better diagnostic performance than the 2017 ICGs in classifying malignant IPMNs.

CLINICAL RELEVANCE STATEMENT

Compared with the radiologists' evaluation based on the 2017 international consensus guidelines, the CT radiomics model exhibited better diagnostic performance in classifying malignant intraductal papillary mucinous neoplasms.

KEY POINTS

• There is a paucity of comparisons between the 2017 international consensus guidelines (ICGs) and radiomics models for malignant intraductal papillary mucinous neoplasms (IPMNs). • The CT radiomics model developed in this study exhibited better diagnostic performance than the 2017 ICGs in classifying malignant IPMNs. • The radiomics model may serve as a valuable complementary tool to the 2017 ICGs, potentially allowing a more quantitative assessment of IPMNs.

Radiomics Based on Contrast-Enhanced Ultrasound Images for Diagnosis of Pancreatic Serous Cystadenoma

OBJECTIVE

The purpose of the study was to develop and validate a radiomics model by using contrast-enhanced ultrasound (CEUS) data for pre-operative differential diagnosis of pancreatic cystic neoplasms (PCNs), especially pancreatic serous cystadenoma (SCA).

METHODS

Patients with pathologically confirmed PCNs who underwent CEUS examination at Chinese PLA hospital from May 2015 to August 2022 were retrospectively collected. Radiomic features were extracted from the regions of interest, which were obtained based on CEUS images. A support vector machine algorithm was used to construct a radiomics model. Moreover, based on the CEUS image features, the CEUS and the combined models were constructed using logistic regression. The performance and clinical utility of the optimal model were evaluated by area under the receiver operating characteristic curve (AUC), sensitivity, specificity and decision curve analysis.

RESULTS

A total of 113 patients were randomly split into the training (n = 79) and test cohorts (n = 34). These patients were pathologically diagnosed with SCA, mucinous cystadenoma, intraductal papillary mucinous neoplasm and solid-pseudopapillary tumor. The radiomics model achieved an AUC of 0.875 and 0.862 in the training and test cohorts, respectively. The sensitivity and specificity of the radiomics model were 81.5% and 86.5% in the training cohort and 81.8% and 91.3% in the test cohort, respectively, which were higher than or comparable with that of the CEUS model and the combined model.

CONCLUSION

The radiomics model based on CEUS images had a favorable differential diagnostic performance in distinguishing SCA from other PCNs, which may be beneficial for the exploration of personalized management strategies.

Preoperative differentiation of pancreatic cystic neoplasm subtypes on computed tomography radiomics

Background

Serous cystic neoplasm (SCN), mucinous cystic neoplasm (MCN), and intraductal papillary mucinous neoplasm (IPMN) comprise a large proportion of pancreatic cystic neoplasms (PCNs). Patients with MCN and IPMN require surgery due to the potential of malignant transformation, whereas those with SCN require periodic surveillance. However, the differential diagnosis of patients with PCNs before treatment remains a great challenge for all surgeons. Therefore, the establishment of a reliable diagnostic tool is urgently required for the improvement of precision diagnostics.

Methods

Between February 2015 and December 2020, 143 consecutive patients with PCNs who were confirmed by postoperative pathology were retrospectively included in the study cohort, then randomized into development and test cohorts at a ratio of 7:3. The predictors of preoperative clinical-radiologic parameters were evaluated by univariate and multivariable logistic regression analyses. A total of 1,218 radiomics features were computationally extracted from the enhanced computed tomography (CT) scans of the tumor region, and a radiomics signature was established by the random forest algorithm. In the development cohort, multi- and binary-class radiomics models integrating preoperative variables and radiomics features were constructed to distinguish between the 3 types of PCNs. The receiver operating characteristic (ROC) curve and the area under the curve (AUC) were used to evaluate the predictive efficiency of the model. An independent internal test cohort was applied to validate the classification models.

Results

All preoperative prediction models were built by integrating the radiomics signature with 13 diagnosis-related radiomics features and 3 important clinical-radiologic parameters: age, sex, and tumor diameter. The multiclass prediction model presented an overall accuracy of 0.804 in the development cohort and 0.707 in the test cohort. The binary-class prediction models displayed higher overall accuracies of 0.853, 0.866, and 0.928 in the development dataset and 0.750, 0.839, and 0.889 in the test dataset. In the test cohort, the binary-class radiomics models showed better predictive performances {AUC =0.914 [95% confidence interval (CI): 0.786 to 1.000], 0.863 (95% CI: 0.714 to 0.941), and 0.926 (95% CI: 0.824 to 1.000)} than the multiclass radiomics model [AUC =0.850 (95% CI: 0.696 to 1.000)], with a large net benefit in the decision curve analysis (DCA). The radiomics-based nomogram provided the correct predicted probability for the diagnosis of PCNs.

Conclusions

The proposed radiomics models with clinical-radiologic parameters and radiomics features help to predict the accurate diagnosis among PCNs to advance personalized medicine.

Adjuvant chemotherapy or no adjuvant chemotherapy? A prediction model for the risk stratification of recurrence or metastasis of nasopharyngeal carcinoma combining MRI radiomics with clinical factors

BACKGROUND

Dose adjuvant chemotherapy (AC) should be offered in nasopharyngeal carcinoma (NPC) patients? Different guidelines provided the different recommendations.

METHODS

In this retrospective study, a total of 140 patients were enrolled and followed for 3 years, with 24 clinical features being collected. The imaging features on the enhanced-MRI sequence were extracted by using PyRadiomics platform. The pearson correlation coefficient and the random forest was used to filter the features associated with recurrence or metastasis. A clinical-radiomics model (CRM) was constructed by the Cox multivariable analysis in training cohort, and was validated in validation cohort. All patients were divided into high- and low-risk groups through the median Rad-score of the model. The Kaplan-Meier survival curves were used to compare the 3-year recurrence or metastasis free rate (RMFR) of patients with or without AC in high- and low-groups.

RESULTS

In total, 960 imaging features were extracted. A CRM was constructed from nine features (seven imaging features and two clinical factors). In the training cohort, the area under curve (AUC) of CRM for 3-year RMFR was 0.872 (P <0.001), and the sensitivity and specificity were 0.935 and 0.672, respectively; In the validation cohort, the AUC was 0.864 (P <0.001), and the sensitivity and specificity were 1.00 and 0.75, respectively. Kaplan-Meier curve showed that the 3-year RMFR and 3-year cancer specific survival (CSS) rate in the high-risk group were significantly lower than those in the low-risk group (P <0.001). In the high-risk group, patients who received AC had greater 3-year RMFR than those who did not receive AC (78.6% vs. 48.1%) (p = 0.03).

CONCLUSION

Considering increasing RMFR, a prediction model for NPC based on two clinical factors and seven imaging features suggested the AC needs to be added to patients in the high-risk group and not in the low-risk group.

The predictive value of a computed tomography-based radiomics model for the surgical separability of thymic epithelial tumors from the superior vena cava and the left innominate vein

Background

The aim of this study was to develop a radiomics machine learning model based on computed tomography (CT) that can predict whether thymic epithelial tumors (TETs) can be separated from veins during surgery and to compare the accuracy of the radiomics model to that of radiologists.

Methods

Patients who underwent thymectomy at our hospital from 2009 to 2017 were included in the screening process. After the selection of patients according to the inclusion and exclusion criteria, the cohort was randomly divided into training and testing groups, and CT images of these patients were collected. Subsequently, two-dimensional (2D) and three-dimensional (3D) regions of interest were labelled using ITK-SNAP 3.8.0 software, and Radiomics features were extracted using Python software (Python Software Foundation) and selected through the least absolute shrinkage and selection operator (LASSO) regression model. To construct the classifier, a support vector machine (SVM) was employed, and a nomogram was created using logistic regression to predict vascular inseparable TETs based on the radiomics score (radscore) and image features. To assess the accuracy of these models, area under receiver operating characteristic (ROC) curves of these models were calculated, and differences among the models were identified using the Delong test.

Results

In this retrospective study, 204 patients with TETs were included, among whom 21 were diagnosed with surgical vascularly inseparable TETs. The area under ROC curve (AUC) of the 2D model, 3D model, 2D + 3D model, and radiologist diagnoses were 0.94, 0.92, 0.95, and 0.87 in the training cohort and 0.95, 0.92, 0.98, and 0.78 in testing cohort, respectively. The Delong test revealed a significant improvement in the performance of the radiomics models compared to radiologists' diagnoses. The logistic regression selected 3 image features, namely maximum diameter of the tumor, degree of abutment of vessel circumference >50%, and absence of the mediastinal fat layer or space between the tumor and surrounding structures. These features, along with the radscore, were included to develop a nomogram. The AUCs of this nomogram were 0.99 in both the training set and testing set, and the Delong test did not find a significant difference between ROC plots of the nomogram and radiomics models.

Conclusions

The proposed radiomics model could accurately predict surgical vascularly inseparable TETs preoperatively and was shown to have a higher predictive value than the radiologists.

Single-cell omics: a new perspective for early detection of pancreatic cancer?

Pancreatic cancer is one of the most lethal cancers, mostly due to late diagnosis and limited treatment options. Early detection of pancreatic cancer in high-risk populations bears the potential to greatly improve outcomes, but current screening approaches remain of limited value despite recent technological advances. This review explores the possible advantages of liquid biopsies for this application, particularly focusing on circulating tumour cells (CTCs) and their subsequent single-cell omics analysis. Originating from both primary and metastatic tumour sites, CTCs provide important information for diagnosis, prognosis and tailoring of treatment strategies. Notably, CTCs have even been detected in the blood of subjects with pancreatic precursor lesions, suggesting their suitability as a non-invasive tool for the early detection of malignant transformation in the pancreas. As intact cells, CTCs offer comprehensive genomic, transcriptomic, epigenetic and proteomic information that can be explored using rapidly developing techniques for analysing individual cells at the molecular level. Studying CTCs during serial sampling and at single-cell resolution will help to dissect tumour heterogeneity for individual patients and among different patients, providing new insights into cancer evolution during disease progression and in response to treatment. Using CTCs for non-invasive tracking of cancer features, including stemness, metastatic potential and expression of immune targets, provides important and readily accessible molecular insights. Finally, the emerging technology of ex vivo culturing of CTCs could create new opportunities to study the functionality of individual cancers at any stage and develop personalised and more effective treatment approaches for this lethal disease.

Machine learning in pancreas surgery, what is new? literature review

Background

Machine learning (ML) is an inquiry domain that aims to establish methodologies that leverage information to enhance performance of various applications. In the healthcare domain, the ML concept has gained prominence over the years. As a result, the adoption of ML algorithms has become expansive. The aim of this scoping review is to evaluate the application of ML in pancreatic surgery.

Methods

We integrated the preferred reporting items for systematic reviews and meta-analyses for scoping reviews. Articles that contained relevant data specializing in ML in pancreas surgery were included.

Results

A search of the following four databases PubMed, Cochrane, EMBASE, and IEEE and files adopted from Google and Google Scholar was 21. The main features of included studies revolved around the year of publication, the country, and the type of article. Additionally, all the included articles were published within January 2019 to May 2022.

Conclusion

The integration of ML in pancreas surgery has gained much attention in previous years. The outcomes derived from this study indicate an extensive literature gap on the topic despite efforts by various researchers. Hence, future studies exploring how pancreas surgeons can apply different learning algorithms to perform essential practices may ultimately improve patient outcomes.

Construction of the XGBoost model for early lung cancer prediction based on metabolic indices

BACKGROUND

Lung cancer is a malignant tumour, and early diagnosis has been shown to improve the survival rate of lung cancer patients. In this study, we assessed the use of plasma metabolites as biomarkers for lung cancer diagnosis. In this work, we used a novel interdisciplinary mechanism, applied for the first time to lung cancer, to detect biomarkers for early lung cancer diagnosis by combining metabolomics and machine learning approaches.

RESULTS

In total, 478 lung cancer patients and 370 subjects with benign lung nodules were enrolled from a hospital in Dalian, Liaoning Province. We selected 47 serum amino acid and carnitine indicators from targeted metabolomics studies using LC‒MS/MS and age and sex demographic indicators of the subjects. After screening by a stepwise regression algorithm, 16 metrics were included. The XGBoost model in the machine learning algorithm showed superior predictive power (AUC = 0.81, accuracy = 75.29%, sensitivity = 74%), with the metabolic biomarkers ornithine and palmitoylcarnitine being potential biomarkers to screen for lung cancer. The machine learning model XGBoost is proposed as an tool for early lung cancer prediction. This study provides strong support for the feasibility of blood-based screening for metabolites and provide a safer, faster and more accurate tool for early diagnosis of lung cancer.

CONCLUSIONS

This study proposes an interdisciplinary approach combining metabolomics with a machine learning model (XGBoost) to predict early the occurrence of lung cancer. The metabolic biomarkers ornithine and palmitoylcarnitine showed significant power for early lung cancer diagnosis.

Update on the Applications of Radiomics in Diagnosis, Staging, and Recurrence of Intrahepatic Cholangiocarcinoma

BACKGROUND

This paper offers an assessment of radiomics tools in the evaluation of intrahepatic cholangiocarcinoma.

METHODS

The PubMed database was searched for papers published in the English language no earlier than October 2022.

RESULTS

We found 236 studies, and 37 satisfied our research criteria. Several studies addressed multidisciplinary topics, especially diagnosis, prognosis, response to therapy, and prediction of staging (TNM) or pathomorphological patterns. In this review, we have covered diagnostic tools developed through machine learning, deep learning, and neural network for the recurrence and prediction of biological characteristics. The majority of the studies were retrospective.

CONCLUSIONS

It is possible to conclude that many performing models have been developed to make differential diagnosis easier for radiologists to predict recurrence and genomic patterns. However, all the studies were retrospective, lacking further external validation in prospective and multicentric cohorts. Furthermore, the radiomics models and the expression of results should be standardized and automatized to be applicable in clinical practice.

Risk stratification for 1- to 2-cm gastric gastrointestinal stromal tumors: visual assessment of CT and EUS high-risk features versus CT radiomics analysis

OBJECTIVES

To investigate the ability of CT and endoscopic sonography (EUS) in predicting the malignant risk of 1-2-cm gastric gastrointestinal stromal tumors (gGISTs) and to clarify whether radiomics could be applied for risk stratification.

METHODS

A total of 151 pathologically confirmed 1-2-cm gGISTs from seven institutions were identified by contrast-enhanced CT scans between January 2010 and March 2021. A detailed description of EUS morphological features was available for 73 gGISTs. The association between EUS or CT high-risk features and pathological malignant potential was evaluated. gGISTs were randomly divided into three groups to build the radiomics model, including 74 in the training cohort, 37 in validation cohort, and 40 in testing cohort. The ROIs covering the whole tumor volume were delineated on the CT images of the portal venous phase. The Pearson test and least absolute shrinkage and selection operator (LASSO) algorithm were used for feature selection, and the ROC curves were used to evaluate the model performance.

RESULTS

The presence of EUS- and CT-based morphological high-risk features, including calcification, necrosis, intratumoral heterogeneity, irregular border, or surface ulceration, did not differ between very-low and intermediate risk 1-2-cm gGISTs (p > 0.05). The radiomics model consisting of five radiomics features showed favorable performance in discrimination of malignant 1-2-cm gGISTs, with the AUC of the training, validation, and testing cohort as 0.866, 0.812, and 0.766, respectively.

CONCLUSIONS

Instead of CT- and EUS-based morphological high-risk features, the CT radiomics model could potentially be applied for preoperative risk stratification of 1-2-cm gGISTs.

KEY POINTS

• The presence of EUS- and CT-based morphological high-risk factors, including calcification, necrosis, intratumoral heterogeneity, irregular border, or surface ulceration, did not correlate with the pathological malignant potential of 1-2-cm gGISTs. • The CT radiomics model could potentially be applied for preoperative risk stratification of 1-2-cm gGISTs.

Clinical impact of artificial intelligence-based solutions on imaging of the pancreas and liver

Artificial intelligence (AI) has experienced substantial progress over the last ten years in many fields of application, including healthcare. In hepatology and pancreatology, major attention to date has been paid to its application to the assisted or even automated interpretation of radiological images, where AI can generate accurate and reproducible imaging diagnosis, reducing the physicians’ workload. AI can provide automatic or semi-automatic segmentation and registration of the liver and pancreatic glands and lesions. Furthermore, using radiomics, AI can introduce new quantitative information which is not visible to the human eye to radiological reports. AI has been applied in the detection and characterization of focal lesions and diffuse diseases of the liver and pancreas, such as neoplasms, chronic hepatic disease, or acute or chronic pancreatitis, among others. These solutions have been applied to different imaging techniques commonly used to diagnose liver and pancreatic diseases, such as ultrasound, endoscopic ultrasonography, computerized tomography (CT), magnetic resonance imaging, and positron emission tomography/CT. However, AI is also applied in this context to many other relevant steps involved in a comprehensive clinical scenario to manage a gastroenterological patient. AI can also be applied to choose the most convenient test prescription, to improve image quality or accelerate its acquisition, and to predict patient prognosis and treatment response. In this review, we summarize the current evidence on the application of AI to hepatic and pancreatic radiology, not only in regard to the interpretation of images, but also to all the steps involved in the radiological workflow in a broader sense. Lastly, we discuss the challenges and future directions of the clinical application of AI methods.

Blood Inflammatory Biomarkers Differentiate Inpatient and Outpatient Coronavirus Disease 2019 From Influenza

Background

The ongoing circulation of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) poses a diagnostic challenge because symptoms of coronavirus disease 2019 (COVID-19) are difficult to distinguish from other respiratory diseases. Our goal was to use statistical analyses and machine learning to identify biomarkers that distinguish patients with COVID-19 from patients with influenza.

Methods

Cytokine levels were analyzed in plasma and serum samples from patients with influenza and COVID-19, which were collected as part of the Centers for Disease Control and Prevention's Hospitalized Adult Influenza Vaccine Effectiveness Network (inpatient network) and the US Flu Vaccine Effectiveness (outpatient network).

Results

We determined that interleukin (IL)-10 family cytokines are significantly different between COVID-19 and influenza patients. The results suggest that the IL-10 family cytokines are a potential diagnostic biomarker to distinguish COVID-19 and influenza infection, especially for inpatients. We also demonstrate that cytokine combinations, consisting of up to 3 cytokines, can distinguish SARS-CoV-2 and influenza infection with high accuracy in both inpatient (area under the receiver operating characteristics curve [AUC] = 0.84) and outpatient (AUC = 0.81) groups, revealing another potential screening tool for SARS-CoV-2 infection.

Conclusions

This study not only reveals prospective screening tools for COVID-19 infections that are independent of polymerase chain reaction testing or clinical condition, but it also emphasizes potential pathways involved in disease pathogenesis that act as potential targets for future mechanistic studies.

Use of Longitudinal Serum Analysis and Machine Learning to Develop a Classifier for Cancer Early Detection

Early detection of solid tumors through a simple screening process, such as the proteomic analysis of biofluids, has the potential to significantly alter the management and outcomes of cancers. The application of advanced targeted proteomics measurements and data analysis strategies to uniformly collected serum or plasma samples would enable longitudinal studies of cancer risk, progression, and response to therapy that have the potential to significantly reduce cancer burden in general. In this article, we describe a generalizable workflow combining robust, multiplexed targeted proteomics measurements applied to longitudinal samples from the Department of Defense Serum Repository with a Random Forest machine learning method for developing and initially evaluating the performance of candidate biomarker panels for early detection of cancers. The effectiveness of this approach was demonstrated in a cohort of 175 head and neck squamous cell carcinoma patients. The outlined protocols include methods for sample preparation, instrument analysis, and data analysis and interpretation using this workflow.

Risk Assessment and Pancreatic Cancer: Diagnostic Management and Artificial Intelligence

Pancreatic cancer (PC) is one of the deadliest cancers, and it is responsible for a number of deaths almost equal to its incidence. The high mortality rate is correlated with several explanations; the main one is the late disease stage at which the majority of patients are diagnosed. Since surgical resection has been recognised as the only curative treatment, a PC diagnosis at the initial stage is believed the main tool to improve survival. Therefore, patient stratification according to familial and genetic risk and the creation of screening protocol by using minimally invasive diagnostic tools would be appropriate. Pancreatic cystic neoplasms (PCNs) are subsets of lesions which deserve special management to avoid overtreatment. The current PC screening programs are based on the annual employment of magnetic resonance imaging with cholangiopancreatography sequences (MR/MRCP) and/or endoscopic ultrasonography (EUS). For patients unfit for MRI, computed tomography (CT) could be proposed, although CT results in lower detection rates, compared to MRI, for small lesions. The actual major limit is the incapacity to detect and characterize the pancreatic intraepithelial neoplasia (PanIN) by EUS and MR/MRCP. The possibility of utilizing artificial intelligence models to evaluate higher-risk patients could favour the diagnosis of these entities, although more data are needed to support the real utility of these applications in the field of screening. For these motives, it would be appropriate to realize screening programs in research settings.

Radiomics: Status quo and future challenges

Noninvasive imaging (computed tomography, magnetic resonance imaging, endoscopic ultrasonography, and positron emission tomography) as an important part of the clinical workflow in the clinic, but it still provides limited information for diagnosis, treatment effect evaluation and prognosis prediction. In addition, judgment and diagnoses made by experts are usually based on multiple years of experience and subjective impression which lead to variable results in the same case. With accumulation of medical imaging data, radiomics emerges as a relatively new approach for analysis. Via artificial intelligence techniques, high-throughput quantitative data which is invisible to the naked eyes extracted from original images can be used in the process of patients’ management. Several studies have evaluated radiomics combined with clinical factors, pathological, or genetic information would assist in the diagnosis, particularly in the prediction of biological characteristics, risk of recurrence, and survival with encouraging results. In various clinical settings, there are limitations and challenges needing to be overcome before transformation. Therefore, we summarize the concepts and method of radiomics including image acquisition, region of interest segmentation, feature extraction and model development. We also set forth the current applications of radiomics in clinical routine. At last, the limitations and related deficiencies of radiomics are pointed out to direct the future opportunities and development.

Differential diagnosis of pancreatic cystic neoplasms through a radiomics-assisted system

Pancreatic cystic neoplasms (PCNs) are a group of heterogeneous diseases with distinct prognosis. Existing differential diagnosis methods require invasive biopsy or prolonged monitoring. We sought to develop an inexpensive, non-invasive differential diagnosis system for PCNs based on radiomics features and clinical characteristics for a higher total PCN screening rate. We retrospectively analyzed computed tomography images and clinical data from 129 patients with PCN, including 47 patients with intraductal papillary mucinous neoplasms (IPMNs), 49 patients with serous cystadenomas (SCNs), and 33 patients with mucinous cystic neoplasms (MCNs). Six clinical characteristics and 944 radiomics features were tested, and nine features were finally selected for model construction using DXScore algorithm. A five-fold cross-validation algorithm and a test group were applied to verify the results. In the five-fold cross-validation section, the AUC value of our model was 0.8687, and the total accuracy rate was 74.23%, wherein the accuracy rates of IPMNs, SCNs, and MCNs were 74.26%, 78.37%, and 68.00%, respectively. In the test group, the AUC value was 0.8462 and the total accuracy rate was 73.61%. In conclusion, our research constructed an end-to-end powerful PCN differential diagnosis system based on radiomics method, which could assist decision-making in clinical practice.

Classification prediction of pancreatic cystic neoplasms based on radiomics deep learning models

BACKGROUND

Preoperative prediction of pancreatic cystic neoplasm (PCN) differentiation has significant value for the implementation of personalized diagnosis and treatment plans. This study aimed to build radiomics deep learning (DL) models using computed tomography (CT) data for the preoperative differential diagnosis of common cystic tumors of the pancreas.

METHODS

Clinical and CT data of 193 patients with PCN were collected for this study. Among these patients, 99 were pathologically diagnosed with pancreatic serous cystadenoma (SCA), 55 were diagnosed with mucinous cystadenoma (MCA) and 39 were diagnosed with intraductal papillary mucinous neoplasm (IPMN). The regions of interest (ROIs) were obtained based on manual image segmentation of CT slices. The radiomics and radiomics-DL models were constructed using support vector machines (SVMs). Moreover, based on the fusion of clinical and radiological features, the best combined feature set was obtained according to the Akaike information criterion (AIC) analysis. Then the fused model was constructed using logistic regression.

RESULTS

For the SCA differential diagnosis, the fused model performed the best and obtained an average area under the curve (AUC) of 0.916. It had a best feature set including position, polycystic features (≥6), cystic wall calcification, pancreatic duct dilatation and radiomics-DL score. For the MCA and IPMN differential diagnosis, the fused model with AUC of 0.973 had a best feature set including age, communication with the pancreatic duct and radiomics score.

CONCLUSIONS

The radiomics, radiomics-DL and fused models based on CT images have a favorable differential diagnostic performance for SCA, MCA and IPMN. These findings may be beneficial for the exploration of individualized management strategies.

Radiomics Combined with Multiple Machine Learning Algorithms in Differentiating Pancreatic Ductal Adenocarcinoma from Pancreatic Neuroendocrine Tumor: More Hands Produce a Stronger Flame

The aim of this study was to assess the diagnostic ability of radiomics combined with multiple machine learning algorithms to differentiate pancreatic ductal adenocarcinoma (PDAC) from pancreatic neuroendocrine tumor (pNET). This retrospective study included a total of 238 patients diagnosed with PDAC or pNET. Using specialized software, radiologists manually mapped regions of interest (ROIs) from computed tomography images and automatically extracted radiomics features. A total of 45 discriminative models were built by five selection algorithms and nine classification algorithms. The performances of the discriminative models were assessed by sensitivity, specificity and the area under receiver operating characteristic curve (AUC) in the training and validation datasets. Using the combination of Gradient Boosting Decision Tree (GBDT) as the selection algorithm and Random Forest (RF) as the classification algorithm, the optimal diagnostic ability with the highest AUC was presented in the training and validation datasets. The sensitivity, specificity and AUC of the model were 0.804, 0.973 and 0.971 in the training dataset and 0.742, 0.934 and 0.930 in the validation dataset, respectively. The combination of radiomics and multiple machine learning algorithms showed the potential ability to discriminate PDAC from pNET. We suggest that multi-algorithm modeling should be considered for similar studies in the future rather than using a single algorithm empirically.

Value assessment of artificial intelligence in medical imaging: a scoping review

BACKGROUND

Artificial intelligence (AI) is seen as one of the major disrupting forces in the future healthcare system. However, the assessment of the value of these new technologies is still unclear, and no agreed international health technology assessment-based guideline exists. This study provides an overview of the available literature in the value assessment of AI in the field of medical imaging.

METHODS

We performed a systematic scoping review of published studies between January 2016 and September 2020 using 10 databases (Medline, Scopus, ProQuest, Google Scholar, and six related databases of grey literature). Information about the context (country, clinical area, and type of study) and mentioned domains with specific outcomes and items were extracted. An existing domain classification, from a European assessment framework, was used as a point of departure, and extracted data were grouped into domains and content analysis of data was performed covering predetermined themes.

RESULTS

Seventy-nine studies were included out of 5890 identified articles. An additional seven studies were identified by searching reference lists, and the analysis was performed on 86 included studies. Eleven domains were identified: (1) health problem and current use of technology, (2) technology aspects, (3) safety assessment, (4) clinical effectiveness, (5) economics, (6) ethical analysis, (7) organisational aspects, (8) patients and social aspects, (9) legal aspects, (10) development of AI algorithm, performance metrics and validation, and (11) other aspects. The frequency of mentioning a domain varied from 20 to 78% within the included papers. Only 15/86 studies were actual assessments of AI technologies. The majority of data were statements from reviews or papers voicing future needs or challenges of AI research, i.e. not actual outcomes of evaluations.

CONCLUSIONS

This review regarding value assessment of AI in medical imaging yielded 86 studies including 11 identified domains. The domain classification based on European assessment framework proved useful and current analysis added one new domain. Included studies had a broad range of essential domains about addressing AI technologies highlighting the importance of domains related to legal and ethical aspects.

Artificial intelligence in pancreatic cancer

Pancreatic cancer is the deadliest disease, with a five-year overall survival rate of just 11%. The pancreatic cancer patients diagnosed with early screening have a median overall survival of nearly ten years, compared with 1.5 years for those not diagnosed with early screening. Therefore, early diagnosis and early treatment of pancreatic cancer are particularly critical. However, as a rare disease, the general screening cost of pancreatic cancer is high, the accuracy of existing tumor markers is not enough, and the efficacy of treatment methods is not exact. In terms of early diagnosis, artificial intelligence technology can quickly locate high-risk groups through medical images, pathological examination, biomarkers, and other aspects, then screening pancreatic cancer lesions early. At the same time, the artificial intelligence algorithm can also be used to predict the survival time, recurrence risk, metastasis, and therapy response which could affect the prognosis. In addition, artificial intelligence is widely used in pancreatic cancer health records, estimating medical imaging parameters, developing computer-aided diagnosis systems, etc. Advances in AI applications for pancreatic cancer will require a concerted effort among clinicians, basic scientists, statisticians, and engineers. Although it has some limitations, it will play an essential role in overcoming pancreatic cancer in the foreseeable future due to its mighty computing power.

Radiomics: a primer on high-throughput image phenotyping

Radiomics is a high-throughput approach to image phenotyping. It uses computer algorithms to extract and analyze a large number of quantitative features from radiological images. These radiomic features collectively describe unique patterns that can serve as digital fingerprints of disease. They may also capture imaging characteristics that are difficult or impossible to characterize by the human eye. The rapid development of this field is motivated by systems biology, facilitated by data analytics, and powered by artificial intelligence. Here, as part of Abdominal Radiology's special issue on Quantitative Imaging, we provide an introduction to the field of radiomics. The technique is formally introduced as an advanced application of data analytics, with illustrating examples in abdominal radiology. Artificial intelligence is then presented as the main driving force of radiomics, and common techniques are defined and briefly compared. The complete step-by-step process of radiomic phenotyping is then broken down into five key phases. Potential pitfalls of each phase are highlighted, and recommendations are provided to reduce sources of variation, non-reproducibility, and error associated with radiomics.

Update on quantitative radiomics of pancreatic tumors

Radiomics is a newer approach for analyzing radiological images obtained from conventional imaging modalities such as computed tomography, magnetic resonance imaging, endoscopic ultrasonography, and positron emission tomography. Radiomics involves extracting quantitative data from the images and assessing them to identify diagnostic or prognostic features such as tumor grade, resectability, tumor response to neoadjuvant therapy, and survival. The purpose of this review is to discuss the basic principles of radiomics and provide an overview of the current clinical applications of radiomics in the field of pancreatic tumors.

A systematic review of radiomics in pancreatitis: applying the evidence level rating tool for promoting clinical transferability

Background

Multiple tools have been applied to radiomics evaluation, while evidence rating tools for this field are still lacking. This study aims to assess the quality of pancreatitis radiomics research and test the feasibility of the evidence level rating tool.

Results

Thirty studies were included after a systematic search of pancreatitis radiomics studies until February 28, 2022, via five databases. Twenty-four studies employed radiomics for diagnostic purposes. The mean ± standard deviation of the adherence rate was 38.3 ± 13.3%, 61.3 ± 11.9%, and 37.1 ± 27.2% for the Radiomics Quality Score (RQS), the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) checklist, and the Image Biomarker Standardization Initiative (IBSI) guideline for preprocessing steps, respectively. The median (range) of RQS was 7.0 (− 3.0 to 18.0). The risk of bias and application concerns were mainly related to the index test according to the modified Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2) tool. The meta-analysis on differential diagnosis of autoimmune pancreatitis versus pancreatic cancer by CT and mass-forming pancreatitis versus pancreatic cancer by MRI showed diagnostic odds ratios (95% confidence intervals) of, respectively, 189.63 (79.65–451.48) and 135.70 (36.17–509.13), both rated as weak evidence mainly due to the insufficient sample size.

Conclusions

More research on prognosis of acute pancreatitis is encouraged. The current pancreatitis radiomics studies have insufficient quality and share common scientific disadvantages. The evidence level rating is feasible and necessary for bringing the field of radiomics from preclinical research area to clinical stage.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13244-022-01279-4.

Radiomics for pseudoprogression prediction in high grade gliomas: added value of MR contrast agent

Objective

Material & methods

Results

Conclusion

•

Radiomics allows for prediction of pseudoprogression in high-grade gliomas.

•

Use of contrast media boosts the performance of the Radiomics prediction model.

Application of Artificial Intelligence in the Management of Pancreatic Cystic Lesions

The rate of incidentally detected pancreatic cystic lesions (PCLs) has increased over the past decade and was recently reported at 8%. These lesions pose a unique challenge, as each subtype of PCL carries a different risk of malignant transformation, ranging from 0% (pancreatic pseudocyst) to 34–68% (main duct intraductal papillary mucinous neoplasm). It is imperative to correctly risk-stratify the malignant potential of these lesions in order to provide the correct care course for the patient, ranging from monitoring to surgical intervention. Even with the multiplicity of guidelines (i.e., the American Gastroenterology Association guidelines and Fukuoka/International Consensus guidelines) and multitude of diagnostic information, risk stratification of PCLs falls short. Studies have reported that 25–64% of patients undergoing PCL resection have pancreatic cysts with no malignant potential, and up to 78% of mucin-producing cysts resected harbor no malignant potential on pathological evaluation. Clinicians are now incorporating artificial intelligence technology to aid in the management of these difficult lesions. This review article focuses on advancements in artificial intelligence within digital pathomics, radiomics, and genomics as they apply to the diagnosis and risk stratification of PCLs.

A deep learning algorithm to improve readers' interpretation and speed of pancreatic cystic lesions on dual-phase enhanced CT

PURPOSE

To develop a deep learning model (DLM) to improve readers' interpretation and speed in the differentiation of pancreatic cystic lesions (PCLs) on dual-phase enhanced CT, and a low contrast media dose, external testing set validated the model.

MATERIALS AND METHODS

Dual-phase enhanced CT images of 363 patients with 368 PCLs obtained from two centers were retrospectively assessed. Based on the examination date, a training and validation set of 266 PCLs, an internal testing set of 52 PCLs were designated from center 1. An external testing set included 50 PCLs from center 2. Clinical and radiological characteristics were compared. The DLM was developed using 3D specially designed densely connected convolutional networks for PCL differentiation. Radiomic features were extracted to build a traditional radiomics model (RM). Performance of the DLM, traditional RM, and three readers was compared.

RESULTS

The accuracy for differential diagnosis was 0.904 with DLM, which was the highest in the internal testing set. Accuracy differences between the DLM and senior radiologist were not significant both in the internal and external testing set (both p > 0.05). With the help of the DLM, the accuracy and specificity of the junior radiologist were significantly improved (all p < 0.05), and all readers' diagnostic time was shortened (all p < 0.05).

CONCLUSION

The DLM achieved senior radiologist-level performance in differentiating benign and malignant PCLs which could improve the junior radiologist's interpretation and speed of PCLs on CT.

A metric learning-based method using graph neural network for pancreatic cystic neoplasm classification from CTs

PURPOSE

Pancreatic cystic neoplasms (PCNs) are relatively rare neoplasms and difficult to be classified preoperatively. Ordinary deep learning methods have great potential to provide support for doctors in PCNs classification but require a quantity of labeled samples and exact segmentation of neoplasm. The proposed metric learning-based method using graph neural network aims to overcome the limitations brought by small and imbalanced dataset and get fast and accurate PCNs classification result from computed tomography (CT) images.

METHODS

The proposed framework applies graph neural network (GNN). GNNs perform well in fusing information and modeling relational data and get better results on dataset with small size. Based on metric learning strategy, model learns distance from the data. The similarity-based algorithm enhances the classification performance, and more characteristic information is found. We use a convolutional neural network (CNN) to extract features from given images. Then GNN is used to find the similarity between each two feature vectors and complete the classification. Several subtasks consisting of randomly selected images are established to improve generalization of the model. The experiments are carried out on the dataset provided by Huashan Hospital. The dataset is labeled by postoperative pathological analysis and contains ROI information calibrated by experts. We set two tasks based on the dataset: benign or malignant diagnosis of PCNs and classification of specific types.

RESULTS

Our model shows good performance on the 2 tasks with accuracies of 88.926% and 74.497%. The comparison of different methods' F1 scores in the benign or malignant diagnosis shows the proposed GNN-based method effectively reduces the negative impact brought by imbalanced dataset, which is also verified by the macro-average comparison in the 4-class classification task.

CONCLUSIONS

Compared with existing models, the proposed GNN-based model shows better performance in terms of imbalanced dataset with small size while reducing labeling cost. The result provides a possibility for its application into the computer aided diagnosis of PCNs. This article is protected by copyright. All rights reserved.

Basic pancreatic lesions: Radiologic-pathologic correlation

The basic pancreatic lesions include location, size, shape, number, capsule, calcification/calculi, hemorrhage, cystic degeneration, fibrosis, pancreatic duct alterations, and microvessel. One or more basic lesions form a kind of pancreatic disease. As recognizing the characteristic imaging features of pancreatic basic lesions and their relationships with pathology aids in differentiating the variety of pancreatic diseases. The purpose of this study is to review the pathological and imaging features of the basic pancreatic lesions.

Cystic Neoplasms of the Pancreas: Differential Diagnosis and Radiology Correlation

Although the probability of pancreatic cystic neoplasms (PCNs) being detected is raising year by year, their differential diagnosis and individualized treatment are still a challenge in clinical work. PCNs are tumors containing cystic components with different biological behaviors, and their clinical manifestations, epidemiology, imaging features, and malignant risks are different. Some are benign [e.g., serous cystic neoplasms (SCNs)], with a barely possible that turning into malignant, while others display a low or higher malignant risk [e.g., solid pseudopapillary neoplasms (SPNs), intraductal papillary mucinous neoplasms (IPMNs), and mucinous cystic neoplasms (MCNs)]. PCN management should concentrate on preventing the progression of malignant tumors while preventing complications caused by unnecessary surgical intervention. Clinically, various advanced imaging equipment are usually combined to obtain a more reliable preoperative diagnosis. The challenge for clinicians and radiologists is how to accurately diagnose PCNs before surgery so that corresponding surgical methods and follow-up strategies can be developed or not, as appropriate. The objective of this review is to sum up the clinical features, imaging findings and management of the most common PCNs according to the classic literature and latest guidelines.

A Clinical Decision Support System for Diabetes Patients with Deep Learning: Experience of a Taiwan Medical Center

Background: Diabetes mellitus (DM) is a major public health problem worldwide. It involves dysfunction of blood sugar regulation resulting from insulin resistance, inadequate insulin secretion, or excessive glucagon secretion. Methods: This study collated 971,401 drug usage records of 51,009 DM patients. These data include patient identification code, age, gender, outpatient visiting dates, visiting code, medication features (included items, doses, and frequencies of drugs), HbA1c results, and testing time. We apply a random forest (RF) model for feature selection and implement a regression model with the bidirectional long short-term memory (Bi-LSTM) deep learning architecture. Finally, we use the root mean square error (RMSE) as the evaluation index for the prediction model. Results: After data cleaning, the data included 8,729 male and 9,115 female cases. Metformin was the most important feature suggested by the RF model, followed by glimepiride, acarbose, pioglitazone, glibenclamide, gliclazide, repaglinide, nateglinide, sitagliptin, and vildagliptin. The model performed better with the past two seasons in the training data than with additional seasons. Further, the Bi-LSTM architecture model performed better than support vector machines (SVMs). Discussion & Conclusion: This study found that Bi-LSTM models is a well kernel in a CDSS which help physicians' decision-making, and the increasing the number of seasons will negative impact the performance. In addition, this study found that the most important drug is metformin, which is recommended as first-line treatment OHA in various situations for DM patients.

AI in spotting high-risk characteristics of medical imaging and molecular pathology

Medical imaging provides a comprehensive perspective and rich information for disease diagnosis. Combined with artificial intelligence technology, medical imaging can be further mined for detailed pathological information. Many studies have shown that the macroscopic imaging characteristics of tumors are closely related to microscopic gene, protein and molecular changes. In order to explore the function of artificial intelligence algorithms in in-depth analysis of medical imaging information, this paper reviews the articles published in recent years from three perspectives: medical imaging analysis method, clinical applications and the development of medical imaging in the direction of pathological molecular prediction. We believe that AI-aided medical imaging analysis will be extensively contributing to precise and efficient clinical decision.

Impact of Interobserver Variability in Manual Segmentation of Non-Small Cell Lung Cancer (NSCLC) Applying Low-Rank Radiomic Representation on Computed Tomography

Simple Summary

Discovery of predictive and prognostic radiomic features in cancer is currently of great interest to the radiologic and oncologic community. Tumor phenotypic and prognostic information can be obtained by extracting features on tumor segmentations, and it is typically imaging analysts, physician trainees, and attending physicians who provide these labeled datasets for analysis. The potential impact of level and type of specialty training on interobserver variability in manual segmentation of NSCLC was examined. Although there was some variability in segmentation between readers, the subsequently extracted radiomic features were overall well correlated. High fidelity radiomic feature extraction relies on accurate feature extraction from imaging that produce robust prognostic and predictive radiomic NSCLC biomarkers. This study concludes that this goal can be obtained using segmenters of different levels of training and clinical experience.

Abstract

This study tackles interobserver variability with respect to specialty training in manual segmentation of non-small cell lung cancer (NSCLC). Four readers included for segmentation are: a data scientist (BY), a medical student (LS), a radiology trainee (MH), and a specialty-trained radiologist (SK) for a total of 293 patients from two publicly available databases. Sørensen–Dice (SD) coefficients and low rank Pearson correlation coefficients (CC) of 429 radiomics were calculated to assess interobserver variability. Cox proportional hazard (CPH) models and Kaplan-Meier (KM) curves of overall survival (OS) prediction for each dataset were also generated. SD and CC for segmentations demonstrated high similarities, yielding, SD: 0.79 and CC: 0.92 (BY-SK), SD: 0.81 and CC: 0.83 (LS-SK), and SD: 0.84 and CC: 0.91 (MH-SK) in average for both databases, respectively. OS through the maximal CPH model for the two datasets yielded c-statistics of 0.7 (95% CI) and 0.69 (95% CI), while adding radiomic and clinical variables (sex, stage/morphological status, and histology) together. KM curves also showed significant discrimination between high- and low-risk patients (p-value < 0.005). This supports that readers’ level of training and clinical experience may not significantly influence the ability to extract accurate radiomic features for NSCLC on CT. This potentially allows flexibility in the training required to produce robust prognostic imaging biomarkers for potential clinical translation.

Kynurenine and Hemoglobin as Sex-Specific Variables in COVID-19 Patients: A Machine Learning and Genetic Algorithms Approach

Differences in clinical manifestations, immune response, metabolic alterations, and outcomes (including disease severity and mortality) between men and women with COVID-19 have been reported since the pandemic outbreak, making it necessary to implement sex-specific biomarkers for disease diagnosis and treatment. This study aimed to identify sex-associated differences in COVID-19 patients by means of a genetic algorithm (GALGO) and machine learning, employing support vector machine (SVM) and logistic regression (LR) for the data analysis. Both algorithms identified kynurenine and hemoglobin as the most important variables to distinguish between men and women with COVID-19. LR and SVM identified C10:1, cough, and lysoPC a 14:0 to discriminate between men with COVID-19 from men without, with LR being the best model. In the case of women with COVID-19 vs. women without, SVM had a higher performance, and both models identified a higher number of variables, including 10:2, lysoPC a C26:0, lysoPC a C28:0, alpha-ketoglutaric acid, lactic acid, cough, fever, anosmia, and dysgeusia. Our results demonstrate that differences in sexes have implications in the diagnosis and outcome of the disease. Further, genetic and machine learning algorithms are useful tools to predict sex-associated differences in COVID-19.

How does the pancreatic solid pseudopapillary neoplasm confuse us: Analyzing from the point view of MRI-based radiomics?

OBJECTIVES

To construct MRI-based radiomics logistic model in differentiating solid pseudopapillary neoplasm (SPN) from three differential diseases containing adenocarcinoma, neuroendocrine tumor (NET), and cystadenoma of pancreas.

MATERIALS AND METHODS

A total of 21 SPNs and 140 differential diseases were enrolled. The MRI images of T1WI, T2WI, DWI, and contrast-enhanced (CE) sequences were taken to delineate the volume of interest, and the corresponding radiomics features were calculated. After the preprocess of data balance and image standardize, the data was divided into training set (6 SPNs and 42 differential diseases) and validation set (15 SPNs and 98 differential diseases) with a proportion of 7:3, randomly. Then after feature selection, four MRI-based logistic models included T1WI, T2WI, DWI, CE, and sum logistic models (Log-T1WI, Log-T2WI, Log-DWI, Log-CE, and Log-sum) were established. The receiver operation curve (ROC) was depicted to evaluate the efficacy of each model.

RESULTS

To the single MRI sequence, the AUCs of Log-T1WI, Log-T2WI, Log-DWI, and Log-CE were similar. Seemingly the AUCs of Log-T2WI were slightly higher with 0. 876 (95%CI, 0.797-0.956) in the training set and 0.853 (95%CI, 0.708-0.998) in the validation set. The Log-sum of four MRI sequences displayed better differentiating efficiency, with AUCs of 0.929 (95%CI, 0.877-0.980) in the training set and 0.925 (95%CI, 0.845-1.000) in the validation set. The Log-Ra/Clin model combined clinical information and radiomics showed the highest AUC of 0.962 (95%CI, 0.919-0.985).

CONCLUSIONS

MRI-based radiomics analysis helped to discern SPNs from radiologically misdiagnosed adenocarcinoma, neuroendocrine tumor, and cystadenoma of pancreas. The efficacy of single sequence logistic model was similar. The Log-sum combined four sequences and Log-Ra/Clin combined clinical information and radiomics demonstrated the better performance in distinction.

A New Dawn for the Use of Artificial Intelligence in Gastroenterology, Hepatology and Pancreatology

Artificial intelligence (AI) is rapidly becoming an essential tool in the medical field as well as in daily life. Recent developments in deep learning, a subfield of AI, have brought remarkable advances in image recognition, which facilitates improvement in the early detection of cancer by endoscopy, ultrasonography, and computed tomography. In addition, AI-assisted big data analysis represents a great step forward for precision medicine. This review provides an overview of AI technology, particularly for gastroenterology, hepatology, and pancreatology, to help clinicians utilize AI in the near future.

Early gastrointestinal cancer: The application of artificial intelligence

Early gastrointestinal (GI) cancer has been the core of clinical endoscopic work. Its early detection and treatment are tightly associated with patients’ prognoses. As a novel technology, artificial intelligence has been improved and applied in the field of endoscopy. Studies on detection, diagnosis, risk, and prognosis evaluation of diseases in the GI tract have been in development, including precancerous lesions, adenoma, early GI cancers, and advanced GI cancers. In this review, research on esophagus, stomach, and colon was concluded, and associated with the process from precancerous lesions to early GI cancer, such as from Barrett’s esophagus to early esophageal cancer, from dysplasia to early gastric cancer, and from adenoma to early colonic cancer. A status quo of research on early GI cancers and artificial intelligence was provided.

CT radiomic features of photodynamic priming in clinical pancreatic adenocarcinoma treatment

Photodynamic therapy (PDT) offers localized focal ablation in unresectable pancreatic tumors while tissues surrounding the treatment volume experience a lower light dose, termed photodynamic priming (PDP). While PDP does not cause tissue damage, it has been demonstrated to promote vascular permeability, improve drug delivery, alleviate tumor cell density, and reduce desmoplasia and the resultant internal pressure in pre-clinical evaluation. Preclinical data supports PDP as a neoadjuvant therapy beneficial to subsequent chemotherapy or immunotherapy, yet it is challenging to quantify PDP effects in clinical treatment without additional imaging and testing. This study investigated the potential of radiomic analysis using CT scans acquired before and after PDT to identify areas experiencing PDT-induced necrosis as well as quantify PDP effects in the surrounding tissues. A total of 235 CT tumor slices from seven patients undergoing PDT for pancreatic tumors were examined. Radiomic features assessed included intensity metrics (CT number in Hounsfield Units) and texture analysis using several gray-level co-occurrence matrix (GLCM) parameters. Pre-treatment scans of tumor areas that resulted in PDT-induced necrosis showed statistically significant differences in intensity and texture-based features that could be used to predict the regions that did respond (paired t-test, response versus no response,p < 0.001). Evaluation of PDP effects on the surrounding tissues also demonstrated statistically significant differences, in tumor mean value, standard deviation, and GLCM parameters of contrast, dissimilarity and homogeneity (t-test, pre versus post,p < 0.001). Using leave-one-out cross validation, six intensity and texture-based features were combined into a support-vector machine model which demonstrated reliable prediction of treatment effects for six out of seven patients (ROC curve, AUC = 0.93). This study provides pilot evidence that texture features extracted from CT scans could be utilized as an effective clinical diagnostic prediction and assessment of PDT and PDP effects in pancreatic tumors. (clinical trial NCT03033225).

Computed tomography-based radiomics approach in pancreatic tumors characterization

Radiomics (or texture analysis) is a new imaging analysis technique that allows calculating the distribution of texture features of pixel and voxel values depend on the type of ROI (3D or 2D), their relationships in the image. Depending on the software, up to several thousand texture elements can be obtained. Radiomics opens up wide opportunities for differential diagnosis and prognosis of pancreatic neoplasias. The aim of this review was to highlight the main diagnostic advantages of texture analysis in different pancreatic tumors. The review describes the diagnostic performance of radiomics in different pancreatic tumor types, application methods, and problems. Texture analysis in PDAC is able to predict tumor grade and associates with lymphovascular invasion and postoperative margin status. In pancreatic neuroendocrine tumors, texture features strongly correlate with differentiation grade and allows distinguishing it from the intrapancreatic accessory spleen. In pancreatic cystic lesions, radiomics is able to accurately differentiate MCN from SCN and distinguish clinically insignificant lesions from IPMNs with advanced neoplasia. In conclusion, the use of the CT radiomics approach provides a higher diagnostic performance of CT imaging in pancreatic tumors differentiation and prognosis. Future studies should be carried out to improve accuracy and facilitate radiomics workflow in pancreatic imaging.

Application of CT-Based Radiomics in Discriminating Pancreatic Cystadenomas From Pancreatic Neuroendocrine Tumors Using Machine Learning Methods

Objectives

The purpose of this study aimed at investigating the reliability of radiomics features extracted from contrast-enhanced CT in differentiating pancreatic cystadenomas from pancreatic neuroendocrine tumors (PNETs) using machine-learning methods.

Methods

In this study, a total number of 120 patients, including 66 pancreatic cystadenomas patients and 54 PNETs patients were enrolled. Forty-eight radiomic features were extracted from contrast-enhanced CT images using LIFEx software. Five feature selection methods were adopted to determine the appropriate features for classifiers. Then, nine machine learning classifiers were employed to build predictive models. The performance of the forty-five models was evaluated with area under the curve (AUC), accuracy, sensitivity, specificity, and F1 score in the testing group.

Results

The predictive models exhibited reliable ability of differentiating pancreatic cystadenomas from PNETs when combined with suitable selection methods. A combination of DC as the selection method and RF as the classifier, as well as Xgboost+RF, demonstrated the best discriminative ability, with the highest AUC of 0.997 in the testing group.

Conclusions

Radiomics-based machine learning methods might be a noninvasive tool to assist in differentiating pancreatic cystadenomas and PNETs.

DeepPrognosis: Preoperative prediction of pancreatic cancer survival and surgical margin via comprehensive understanding of dynamic contrast-enhanced CT imaging and tumor-vascular contact parsing

Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal cancers and carries a dismal prognosis of ∼10% in five year survival rate. Surgery remains the best option of a potential cure for patients who are evaluated to be eligible for initial resection of PDAC. However, outcomes vary significantly even among the resected patients who were the same cancer stage and received similar treatments. Accurate quantitative preoperative prediction of primary resectable PDACs for personalized cancer treatment is thus highly desired. Nevertheless, there are a very few automated methods yet to fully exploit the contrast-enhanced computed tomography (CE-CT) imaging for PDAC prognosis assessment. CE-CT plays a critical role in PDAC staging and resectability evaluation. In this work, we propose a novel deep neural network model for the survival prediction of primary resectable PDAC patients, named as 3D Contrast-Enhanced Convolutional Long Short-Term Memory network (CE-ConvLSTM), which can derive the tumor attenuation signatures or patterns from patient CE-CT imaging studies. Tumor-vascular relationships, which might indicate the resection margin status, have also been proven to hold strong relationships with the overall survival of PDAC patients. To capture such relationships, we propose a self-learning approach for automated pancreas and peripancreatic anatomy segmentation without requiring any annotations on our PDAC datasets. We then employ a multi-task convolutional neural network (CNN) to accomplish both tasks of survival outcome and margin prediction where the network benefits from learning the resection margin related image features to improve the survival prediction. Our presented framework can improve overall survival prediction performances compared with existing state-of-the-art survival analysis approaches. The new staging biomarker integrating both the proposed risk signature and margin prediction has evidently added values to be combined with the current clinical staging system.

Radiomics Analysis of Contrast-Enhanced CT Predicts Survival in Clear Cell Renal Cell Carcinoma

Purpose

To develop and validate the radiomics nomogram that combines clinical factors and radiomics features to estimate overall survival (OS) in patients with clear cell renal cell carcinoma (ccRCC), and assess the incremental value of radiomics for OS estimation.

Materials and Methods

One hundred ninety-four ccRCC cases were included in the training cohort and 188 ccRCC patients from another hospital as the test cohort. Three-dimensional region-of-interest segmentation was manually segmented on multiphasic contrast-enhanced abdominal CT images. Radiomics score (Rad-score) was calculated from a formula generated via least absolute shrinkage and selection operator (LASSO) Cox regression, after which the association between the Rad-score and OS was explored. The radiomics nomogram (clinical factors + Rad-score) was developed to demonstrate the incremental value of the Rad-score to the clinical nomogram for individualized OS estimation, which was then evaluated in relation to calibration and discrimination.

Results

Rad-score, calculated using a linear combination of the 11 screened features multiplied by their respective LASSO Cox coefficients, was significantly associated with OS. Calibration curves showed good agreement between the OS predicted by the nomograms and observed outcomes. The radiomics nomogram presented higher discrimination capability compared to clinical nomogram in the training (C-index: 0.884; 95% CI: 0.808–0.940 vs. 0.803; 95% CI: 0.705–0.899, P < 0.05) and test cohorts (C-index: 0.859; 95% CI: 0.800–0.921 vs. 0.846; 95% CI: 0.777–0.915, P < 0.05).

Conclusions

The radiomics nomogram may be used for predicting OS in patients with ccRCC, and radiomics is useful to assist quantitative and personalized treatment.

CT texture analysis of abdominal lesions – Part II: Tumors of the Kidney and Pancreas

It has been proven in a few early studies that radiomic analysis offers a promising opportunity to detect or differentiate between organ lesions based on their unique texture parameters. Recently, the utilization of CT texture analysis (CTTA) has been receiving significant attention, especially for response evaluation and prognostication of different oncological diagnoses. In this review article, we discuss the unique ability of radiomics and its subfield CTTA to diagnose lesions in the pancreas and kidney. We review studies in which CTTA was used for the classification of histology grades in pancreas and kidney tumors. We also review the role of radiogenomics in the prediction of the molecular and genetic subtypes of pancreatic tumors. Furthermore, we provide a short report on recent advancements of radiomic analysis in predicting prognosis and survival of patients with pancreatic and renal cancers.